| description abstract | The design of manufacturing systems is typically decoupled from ordering policy decisions. Traditionally, the system design decision is made to minimize the total investment cost given some system capacity requirements. Then, after the system is implemented, the ordering policy decisions are made. In this paper, a coupled approach is presented whereby the manufacturing system design is jointly developed with the ordering policy to minimize the total cost of inventory holding, setup, and equipment investment in a multiproduct system. The methodology is presented in the context of scalable-reconfigurable manufacturing systems (scalable-RMSs). First, a linear integer mathematical formulation to minimize investment cost in a single-product, multistage scalable-RMS is presented. The mathematical formulation is then extended to consider multiple products. Due to the nonlinear nature of the multiproduct formulation, an iterative algorithm is developed. Lastly, a mathematical formulation to simultaneously minimize the system investment and operating costs (i.e., the coupled approach) is presented. Given the complexity of the formulation, a genetic algorithm (GA)-based heuristic is proposed. Twenty four instances of the problem were generated to test the proposed methodologies. Experimental results indicate that the proposed GA-based heuristic is efficient in terms of solution quality and runtime. Further, experimental results indicate that the proposed coupled approach reduces the total costs by an average of 25% over the decoupled approach. It is concluded that the coupled approach (solved with the proposed GA-based heuristic) outperforms the decoupled approach (solved to optimality) in all instances considered. | |
